LiteCON: An All-Photonic Neuromorphic Accelerator for Energy-efficient Deep Learning (Preprint)
This work addresses the problem of high energy consumption in deep learning training and inference for applications requiring efficient hardware accelerators, representing a novel approach rather than an incremental improvement.
The paper tackles the challenge of energy-efficient deep learning by proposing LiteCON, an all-photonic neuromorphic accelerator that improves CNN throughput by up to 32x and energy efficiency by up to 37x with minimal accuracy loss.
Deep learning is highly pervasive in today's data-intensive era. In particular, convolutional neural networks (CNNs) are being widely adopted in a variety of fields for superior accuracy. However, computing deep CNNs on traditional CPUs and GPUs brings several performance and energy pitfalls. Several novel approaches based on ASIC, FPGA, and resistive-memory devices have been recently demonstrated with promising results. Most of them target only the inference (testing) phase of deep learning. There have been very limited attempts to design a full-fledged deep learning accelerator capable of both training and inference. It is due to the highly compute and memory-intensive nature of the training phase. In this paper, we propose LiteCON, a novel analog photonics CNN accelerator. LiteCON uses silicon microdisk-based convolution, memristor-based memory, and dense-wavelength-division-multiplexing for energy-efficient and ultrafast deep learning. We evaluate LiteCON using a commercial CAD framework (IPKISS) on deep learning benchmark models including LeNet and VGG-Net. Compared to the state-of-the-art, LiteCON improves the CNN throughput, energy efficiency, and computational efficiency by up to 32x, 37x, and 5x respectively with trivial accuracy degradation.